Alternative methods of selection for litter size in mice. III. Response to 21 generations of selection

Alternative methods of selection to increase litter size in mice have been practiced for 21 generations followed by six generations of relaxed selection. Three replicates were used with four selection criteria: index of components (IX: I = 1.21 X total ovulation rate + 9.05 X ova success), uterine capacity (UT), litter size (LS), and an unselected control (LC). In IX, ovulation rate and ova success were measured by number of corpora lutea and number of pups born/number of corpora lutea, respectively. In UT, uterine capacity was measured and defined as number of pups born to unilaterally ovariectomized (right ovary excised) females. Selection in LS was based on number born to unaltered dams. In all cases, number born was fully formed, live or dead pups. Pups from 16 randomly chosen LC dams and from the top 16 dams in IX, UT, and LS were selected to produce the next generation in each criterion-replicate line. Response in number born, selected criteria deviated from control, was regressed on generation number over the 21 generations of selection. Responses for the IX and LS criteria were quite similar (.14 +/- .01 and .16 +/- .01 pups per generation, respectively), whereas response in UT, with only one functional horn, was slightly lower (.09 +/- .01). The average cumulative selection differentials for IX, LS, and UT at Generation 21 were 32.78 index units, 36.38 pups, and 28.53 pups, respectively. The LC criterion had an unintentional cumulative selection differential of 3.3 pups. Regressions of response in number born (litter size in IX and LS, uterine capacity in UT) on cumulative selection differential for IX, LS, and UT were .10 +/- .01, .10 +/- .01, and .08 +/- .01, respectively. During the relaxed selection, mean responses for litter size in IX, LS, and UT criteria were 3.17, 4.09, and 1.67 pups, respectively.